5 files changed, 98 insertions, 26 deletions

diff --git a/arch/powerpc/platforms/powernv/memtrace.c b/arch/powerpc/platforms/powernv/memtrace.c
index cfbd242c3e01..bff8e38a7e9a 100644
--- a/arch/powerpc/platforms/powernv/memtrace.c
+++ b/arch/powerpc/platforms/powernv/memtrace.c
@@ -99,6 +99,23 @@ static int change_memblock_state(struct memory_block *mem, void *arg)
 	return 0;
 }
 
+static void memtrace_clear_range(unsigned long start_pfn,
+				 unsigned long nr_pages)
+{
+	unsigned long pfn;
+
+	/*
+	 * As pages are offline, we cannot trust the memmap anymore. As HIGHMEM
+	 * does not apply, avoid passing around "struct page" and use
+	 * clear_page() instead directly.
+	 */
+	for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) {
+		if (IS_ALIGNED(pfn, PAGES_PER_SECTION))
+			cond_resched();
+		clear_page(__va(PFN_PHYS(pfn)));
+	}
+}
+
 /* called with device_hotplug_lock held */
 static bool memtrace_offline_pages(u32 nid, u64 start_pfn, u64 nr_pages)
 {
@@ -144,6 +161,11 @@ static u64 memtrace_alloc_node(u32 nid, u64 size)
 	for (base_pfn = end_pfn; base_pfn > start_pfn; base_pfn -= nr_pages) {
 		if (memtrace_offline_pages(nid, base_pfn, nr_pages) == true) {
 			/*
+			 * Clear the range while we still have a linear
+			 * mapping.
+			 */
+			memtrace_clear_range(base_pfn, nr_pages);
+			/*
 			 * Remove memory in memory block size chunks so that
 			 * iomem resources are always split to the same size and
 			 * we never try to remove memory that spans two iomem
diff --git a/arch/powerpc/platforms/powernv/opal-dump.c b/arch/powerpc/platforms/powernv/opal-dump.c
index 4c827826c05e..1a8b6e276a11 100644
--- a/arch/powerpc/platforms/powernv/opal-dump.c
+++ b/arch/powerpc/platforms/powernv/opal-dump.c
@@ -319,15 +319,14 @@ static ssize_t dump_attr_read(struct file *filep, struct kobject *kobj,
 	return count;
 }
 
-static struct dump_obj *create_dump_obj(uint32_t id, size_t size,
-					uint32_t type)
+static void create_dump_obj(uint32_t id, size_t size, uint32_t type)
 {
 	struct dump_obj *dump;
 	int rc;
 
 	dump = kzalloc(sizeof(*dump), GFP_KERNEL);
 	if (!dump)
-		return NULL;
+		return;
 
 	dump->kobj.kset = dump_kset;
 
@@ -347,34 +346,51 @@ static struct dump_obj *create_dump_obj(uint32_t id, size_t size,
 	rc = kobject_add(&dump->kobj, NULL, "0x%x-0x%x", type, id);
 	if (rc) {
 		kobject_put(&dump->kobj);
-		return NULL;
+		return;
 	}
 
+	/*
+	 * As soon as the sysfs file for this dump is created/activated there is
+	 * a chance the opal_errd daemon (or any userspace) might read and
+	 * acknowledge the dump before kobject_uevent() is called. If that
+	 * happens then there is a potential race between
+	 * dump_ack_store->kobject_put() and kobject_uevent() which leads to a
+	 * use-after-free of a kernfs object resulting in a kernel crash.
+	 *
+	 * To avoid that, we need to take a reference on behalf of the bin file,
+	 * so that our reference remains valid while we call kobject_uevent().
+	 * We then drop our reference before exiting the function, leaving the
+	 * bin file to drop the last reference (if it hasn't already).
+	 */
+
+	/* Take a reference for the bin file */
+	kobject_get(&dump->kobj);
 	rc = sysfs_create_bin_file(&dump->kobj, &dump->dump_attr);
-	if (rc) {
+	if (rc == 0) {
+		kobject_uevent(&dump->kobj, KOBJ_ADD);
+
+		pr_info("%s: New platform dump. ID = 0x%x Size %u\n",
+			__func__, dump->id, dump->size);
+	} else {
+		/* Drop reference count taken for bin file */
 		kobject_put(&dump->kobj);
-		return NULL;
 	}
 
-	pr_info("%s: New platform dump. ID = 0x%x Size %u\n",
-		__func__, dump->id, dump->size);
-
-	kobject_uevent(&dump->kobj, KOBJ_ADD);
-
-	return dump;
+	/* Drop our reference */
+	kobject_put(&dump->kobj);
+	return;
 }
 
 static irqreturn_t process_dump(int irq, void *data)
 {
 	int rc;
 	uint32_t dump_id, dump_size, dump_type;
-	struct dump_obj *dump;
 	char name[22];
 	struct kobject *kobj;
 
 	rc = dump_read_info(&dump_id, &dump_size, &dump_type);
 	if (rc != OPAL_SUCCESS)
-		return rc;
+		return IRQ_HANDLED;
 
 	sprintf(name, "0x%x-0x%x", dump_type, dump_id);
 
@@ -386,12 +402,10 @@ static irqreturn_t process_dump(int irq, void *data)
 	if (kobj) {
 		/* Drop reference added by kset_find_obj() */
 		kobject_put(kobj);
-		return 0;
+		return IRQ_HANDLED;
 	}
 
-	dump = create_dump_obj(dump_id, dump_size, dump_type);
-	if (!dump)
-		return -1;
+	create_dump_obj(dump_id, dump_size, dump_type);
 
 	return IRQ_HANDLED;
 }
diff --git a/arch/powerpc/platforms/powernv/opal-elog.c b/arch/powerpc/platforms/powernv/opal-elog.c
index ecd6d9177d13..4113b19059d8 100644
--- a/arch/powerpc/platforms/powernv/opal-elog.c
+++ b/arch/powerpc/platforms/powernv/opal-elog.c
@@ -183,14 +183,14 @@ static ssize_t raw_attr_read(struct file *filep, struct kobject *kobj,
 	return count;
 }
 
-static struct elog_obj *create_elog_obj(uint64_t id, size_t size, uint64_t type)
+static void create_elog_obj(uint64_t id, size_t size, uint64_t type)
 {
 	struct elog_obj *elog;
 	int rc;
 
 	elog = kzalloc(sizeof(*elog), GFP_KERNEL);
 	if (!elog)
-		return NULL;
+		return;
 
 	elog->kobj.kset = elog_kset;
 
@@ -223,18 +223,37 @@ static struct elog_obj *create_elog_obj(uint64_t id, size_t size, uint64_t type)
 	rc = kobject_add(&elog->kobj, NULL, "0x%llx", id);
 	if (rc) {
 		kobject_put(&elog->kobj);
-		return NULL;
+		return;
 	}
 
+	/*
+	 * As soon as the sysfs file for this elog is created/activated there is
+	 * a chance the opal_errd daemon (or any userspace) might read and
+	 * acknowledge the elog before kobject_uevent() is called. If that
+	 * happens then there is a potential race between
+	 * elog_ack_store->kobject_put() and kobject_uevent() which leads to a
+	 * use-after-free of a kernfs object resulting in a kernel crash.
+	 *
+	 * To avoid that, we need to take a reference on behalf of the bin file,
+	 * so that our reference remains valid while we call kobject_uevent().
+	 * We then drop our reference before exiting the function, leaving the
+	 * bin file to drop the last reference (if it hasn't already).
+	 */
+
+	/* Take a reference for the bin file */
+	kobject_get(&elog->kobj);
 	rc = sysfs_create_bin_file(&elog->kobj, &elog->raw_attr);
-	if (rc) {
+	if (rc == 0) {
+		kobject_uevent(&elog->kobj, KOBJ_ADD);
+	} else {
+		/* Drop the reference taken for the bin file */
 		kobject_put(&elog->kobj);
-		return NULL;
 	}
 
-	kobject_uevent(&elog->kobj, KOBJ_ADD);
+	/* Drop our reference */
+	kobject_put(&elog->kobj);
 
-	return elog;
+	return;
 }
 
 static irqreturn_t elog_event(int irq, void *data)
diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c
index 888aa9584e94..0693fd16e2c9 100644
--- a/arch/powerpc/platforms/powernv/setup.c
+++ b/arch/powerpc/platforms/powernv/setup.c
@@ -124,12 +124,29 @@ static void pnv_setup_rfi_flush(void)
 			type = L1D_FLUSH_ORI;
 	}
 
+	/*
+	 * If we are non-Power9 bare metal, we don't need to flush on kernel
+	 * entry or after user access: they fix a P9 specific vulnerability.
+	 */
+	if (!pvr_version_is(PVR_POWER9)) {
+		security_ftr_clear(SEC_FTR_L1D_FLUSH_ENTRY);
+		security_ftr_clear(SEC_FTR_L1D_FLUSH_UACCESS);
+	}
+
 	enable = security_ftr_enabled(SEC_FTR_FAVOUR_SECURITY) && \
 		 (security_ftr_enabled(SEC_FTR_L1D_FLUSH_PR)   || \
 		  security_ftr_enabled(SEC_FTR_L1D_FLUSH_HV));
 
 	setup_rfi_flush(type, enable);
 	setup_count_cache_flush();
+
+	enable = security_ftr_enabled(SEC_FTR_FAVOUR_SECURITY) &&
+		 security_ftr_enabled(SEC_FTR_L1D_FLUSH_ENTRY);
+	setup_entry_flush(enable);
+
+	enable = security_ftr_enabled(SEC_FTR_FAVOUR_SECURITY) &&
+		 security_ftr_enabled(SEC_FTR_L1D_FLUSH_UACCESS);
+	setup_uaccess_flush(enable);
 }
 
 static void __init pnv_setup_arch(void)
diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c
index c17f81e433f7..11d8fde770c3 100644
--- a/arch/powerpc/platforms/powernv/smp.c
+++ b/arch/powerpc/platforms/powernv/smp.c
@@ -44,7 +44,7 @@
 #include <asm/udbg.h>
 #define DBG(fmt...) udbg_printf(fmt)
 #else
-#define DBG(fmt...)
+#define DBG(fmt...) do { } while (0)
 #endif
 
 static void pnv_smp_setup_cpu(int cpu)